HAL Id: hal-02310350
https://hal.archives-ouvertes.fr/hal-02310350
Submitted on 27 Feb 2020
HAL is a multi-disciplinary open access
archive for the deposit and dissemination of
sci-entific research documents, whether they are
pub-lished or not. The documents may come from
teaching and research institutions in France or
abroad, or from public or private research centers.
L’archive ouverte pluridisciplinaire HAL, est
destinée au dépôt et à la diffusion de documents
scientifiques de niveau recherche, publiés ou non,
émanant des établissements d’enseignement et de
recherche français ou étrangers, des laboratoires
publics ou privés.
Distributed under a Creative Commons Attribution| 4.0 International License
Management of liver failure in general intensive care unit
C. Paugam-Burtz, E. Levesque, A. Louvet, D. Thabut, R. Amathieu, C.
Bureau, C. Camus, Gerald Chanques, S. Faure, M. Ferrandière, et al.
To cite this version:
C. Paugam-Burtz, E. Levesque, A. Louvet, D. Thabut, R. Amathieu, et al.. Management of liver
failure in general intensive care unit. Anaesthesia Critical Care & Pain Medicine, Elsevier Masson,
2020, 39 (1), pp.143-161. �10.1016/j.accpm.2019.06.014�. �hal-02310350�
Guidelines
Management
of
liver
failure
in
general
intensive
care
unit
§,§§
C.
Paugam-Burtz
1,2,
E.
Levesque
3,4,
A.
Louvet
5,
D.
Thabut
6,
R.
Amathieu
7,8,
C.
Bureau
9,10,11,
C.
Camus
12,
G.
Chanques
13,
S.
Faure
14,
M.
Ferrandie`re
15,
C.
Francoz
16,17,
A.
Galbois
18,
T.
Gustot
19,20,
C.
Ichai
21,
P.
Ichai
22,23,24,
S.
Jaber
25,
T.
Lescot
26,
R.
Moreau
27,28,29,30,
S.
Roullet
31,32,
F.
Saliba
33,
T.
The´venot
34,
L.
Velly
35,36,
E.
Weiss
37,38,*
1
DepartmentofAnaesthesiologyandCriticalCare,BeaujonHospital,DMUParabol,AP–HPNord,Universite´ deParis,92110Clichy,France
2
UMR_S1149,Centrederecherchesurl’inflammation,Inserm,Universite´ deParis,France
3
DepartmentofAnaesthesiologyandCriticalCare,Henri-MondorHospital,AssistancePublique–HoˆpitauxdeParis,94010Cre´teil,France
4
EADynamycUPEC,ENVAFaculte´ deMe´decinedeCre´teil,94000Cre´teil,France
5
DepartmentofDigestiveDiseases,ClaudeHuriezHospital,59037Lille,France
6DepartmentofHepathology,Pitie´-Salpeˆtrie`reHospital,Assistancepublique–HoˆpitauxdeParis,75013Paris,France 7Universite´ Paris13–UFRSMBH–CNRSUMR7244,Paris,France
8
CriticalCareDepartment,DiaconessesCroixSaintSimonHospitalGroup,75020Paris,France
9
DepartmentofHepathology,PurpanHospital,ToulouseUniversityHospital,31300ToulouseFrance
10
InstitutCardiomet,CardiovascularandMetabolicDepartment,RangueilHospital,TSA50032,31059Toulousecedex9,France
11
PaulSabatierUniversity,31330Toulouse,France
12
DepartmentofInfectiousDiseasesandMedicalResuscitation,RennesUniversityHospital,35000Rennes,France
13DepartmentofAnaesthesiaandIntensiveCare,SaintEloiMontpellierUniversityHospital,andPhyMedExp,UniversityofMontpellier,INSERM,CNRS,
34295Montpelliercedex5,France
14
DepartmentofHepathology,SaintEloiMontpellierUniversityHospital,34090Montpellier,France
15
SurgicalIntensiveCareUnit,ToursUniversityHospital,37044Tourscedex9,France
16
DepartmentofHepathology,BeaujonHospital,AP–HP,92110Clichy,France
17
UMR_S1149,Centrederecherchesurl’inflammation,INSERMandParisDiderotUniversity,75013Paris,France
18
RamsayGe´ne´raledeSante´,ClaudeGalienPrivateHospital,DepartmentofPolyvalentResuscitation,91480Quincy-sous-Se´nart,France
19DepartmentofGastroenterologyandHepato-Pancreatology,ErasmeHospital,RoutedeLennik,808,1070Bruxelles,Belgium 20
LaboratoryofExperimentalGastroenterology,Faculte´ deMe´decine,Universite´ LibredeBruxelles,avenueFranklin-Roosevelt,50,1050Bruxelles,Belgium
21
UniversityofCoˆteD’Azur,NiceUniversityHospital,DepartmentofPolyvalentResuscitation,Pasteur2Hospital,06000Nice,France
22
Hepato-BiliaryCenter,Paul-BrousseHospital,AP–HP,LiverIntensiveCareUnit,94800Villejuif,France
23
INSERM,Unite´ 1193,Universite´ Paris-Saclay,94800Villejuif,France
24
DHUHepatinov,94800Villejuif,France
25DepartmentofAnaesthesiaandIntensiveCare,SaintEloiMontpellierUniversityHospital,andPhyMedExp,UniversityofMontpellier,INSERM,CNRS,
34295Montpelliercedex5,France
26
SorbonneUniversite´,DepartmentofAnesthesiologyandCriticalCareMedicine,Saint-AntoineHospital,AssistancePublique-HoˆpitauxdeParis,ParisFrance
27
Inserm,Universite´ ParisDiderot,CentredeRecherchesurl’Inflammation(CRI),75018Paris,France
28
DepartmentofHepatology,BeaujonHospital,AP–HP,92110Clichy,France
29
EuropeanFoundationforthestudyonchronicliverfailure(EFClif),Barcelona,Spain
30
InstituteforLiverandBiliarySciences(ILBS),NewDelhi,India
31AnaesthesiologyandCriticalCareDepartment1,BordeauxUniversityHospital,33000Bordeaux,France 32UniversityofBordeaux,INSERMU1034,BiologyofCardiovascularDiseases,33000Bordeaux,France 33
Hepato-BiliaryCenter,PaulBrousseHospital,AP-HP,INSERMUnite´ 935andUnite´ 1193,94800Villejuif,France
34
DepartmentofHepatologyandDigestiveCriticalCare,DepartmentofHepatology,JeanMinjozUniversityHospital,25030Besanc¸onFrance
35
AixMarseilleUniversity,CNRS,InstitutdeNeurosciencedelaTimone(INT),13005Marseille,France
36
DepartmentofAnaesthesiologyandCriticalCareMedicine,UniversityHospitalLaTimone,13005Marseille,France
37DepartmentofAnesthesiologyandCriticalCare,BeaujonHospital,DMUParabol,AP–HPNord,UniversityofParis,France 38UMR_S1149,Centrederecherchesurl’inflammation,INSERMandUniversityofParis,Paris,France
§
ValidatedbytheSFARCouncilonJune21st,2018.
§§
Withthecollaborationofthefollowingsocieties:Socie´te´ franc¸aised’anesthe´sieetdere´animation,Associationfranc¸aisepourl’e´tudedufoie. * Correspondingauthorat:Serviced’anesthe´sie-re´animation,hoˆpitalBeaujon,100,boulevardduGe´ne´ral-Leclerc,92110Clichy,France.
E-mailaddress:emmanuel.weiss@aphp.fr(E.Weiss).
https://doi.org/10.1016/j.accpm.2019.06.014
2352-5568/ C2019TheAuthors.PublishedbyElsevierMassonSASonbehalfofSocie´te´ franc¸aised’anesthe´sieetdere´animation(Sfar).Thisisanopenaccessarticleunderthe
Workinggroups
French Society of Anesthesiology and Critical Care Medicine (Socie´te´ Franc¸aised’Anesthe´sie-Re´animation,SFAR)andtheFrench AssociationfortheStudy oftheLiver(Associationfranc¸aisepour l’e´tudedufoie,AFEF)
Experts’coordinators C.Paugam-Burtz E.Levesque A.Louvet D.Thabut Organisers E.Weiss L.Velly Experts R.Amathieu C.Bureau C.Camus G.Chanques S.Faure M.Ferrandie`re C.Francoz A.Galbois T.Gustot P.Ichai C.Ichai S.Jaber T.Lescot R.Moreau S.Roullet F.Saliba T.Thevenot 1. Introduction
Liverfailureisaconditionthatcanbemanagedormayoccurin generalcriticalcareundertwoverydifferentcircumstances:onthe one hand, acute liver failure (ALF) corresponding to a sudden deterioration inliver functiondue totoxic or infectiousagents occurringintheabsenceofunderlyingliverdisease.AlthoughALF isuncommon,itismandatorytorecogniseitatanearlystageso thatafirst-linediagnosticandtherapeuticstrategycanbeadopted, andexpertadvicesought.Ontheotherhand,thehospitalisation in critical care of cirrhotic patients is common, either for the complications of cirrhosis itself or for medical events likely to decompensatepreexistingliverdisease.
ARTICLE INFO
Articlehistory:
Availableonline13September2019
Keywords: Acuteliverfailure
Acute-on-chronicliverfailure Guidelines
Cirrhosis Intensivecareunit
ABSTRACT
Objective:ToproduceFrenchguidelinesonManagementofLiverfailureingeneralIntensiveCareUnit (ICU).
Design:Aconsensuscommitteeof23expertsfromtheFrenchSocietyofAnesthesiologyandCriticalCare Medicine(Socie´te´ franc¸aised’anesthe´sieetdere´animation,SFAR)andtheFrenchAssociationforthe StudyoftheLiver(Associationfranc¸aisepourl’e´tudedufoie,AFEF)wasconvened.Aformal conflict-of-interest(COI)policywasdevelopedatthestartoftheprocessandenforcedthroughout.Theentire guidelineprocesswasconductedindependentlyofanyindustrialfunding.Theauthorswereadvisedto followtheprinciplesoftheGradingofRecommendationsAssessment,DevelopmentandEvaluation (GRADE)systemtoguidetheirassessmentofthequalityofevidence.Thepotentialdrawbacksofmaking strongrecommendationsin thepresenceof low-quality evidencewereemphasised. Some recom-mendationswereungraded.
Methods:Twofieldsweredefined:acuteliverfailure(ALF)andcirrhoticpatientsingeneralICU.The panelfocusedonthree questionswith respect to ALF:(1)Whichetiological examinations should beperformedtoreducemorbidityandmortality?(2)Whichspecifictreatmentsshouldbeinitiated rapidlytoreducemorbidityandmortality?(3)Whichsymptomatictreatmentshouldbeinitiatedrapidly toreducemorbidityandmortality?Sevenquestionsconcerningcirrhoticpatientswereaddressed: (1)WhichcriteriashouldbeusedtoguideICUadmissionofcirrhoticpatientsinordertoimprovetheir prognosis?(2)Whichspecificmanagementofkidneyinjuryshouldbeimplementedtoreducemorbidity andmortalityincirrhoticICUpatients?(3)Whichspecificmeasurestomanagesepsisinordertoreduce morbidityandmortalityincirrhoticICUpatients?(4)Inwhichcircumstances,humanserumalbumin shouldbeadministeredtoreducemorbidityandmortalityincirrhoticICUpatients?(5)Howshould digestivehaemorrhagebetreatedinordertoreducemorbidityandmortalityincirrhoticICUpatients? (6)HowshouldhaemostasisbemanagedinordertoreducemorbidityandmortalityincirrhoticICU patients?And(7)Whenshouldadvicebeobtainedfromanexpertcentreinordertoreducemorbidity andmortality in cirrhoticICU patients?Population, intervention,comparisonand outcome (PICO) issueswerereviewedandupdatedasrequired,andevidenceprofilesweregenerated.Ananalysisofthe literatureandrecommendationswasthenperformedinaccordancewiththeGRADE1
methodology. Results:TheSFAR/AFEFGuidelinespanelproduced18statementsonliverfailureingeneralICU.Aftertwo rounds of debate and various amendments, a strong agreement was reached on 100% of the recommendations:sixhadahighlevelofevidence(Grade1),sevenhadalowlevelofevidence(Grade 2)andsixwereexpertjudgments.Finally,norecommendationwasprovidedwithrespecttoonequestion. Conclusions:Substantialagreementexistsamongexpertsregardingnumerousstrongrecommendations ontheoptimumcareofpatientswithliverfailureingeneralICU.
C 2019TheAuthors.PublishedbyElsevierMassonSASonbehalfofSocie´te´ franc¸aised’anesthe´sieetde
re´animation(Sfar).ThisisanopenaccessarticleundertheCCBYlicense(http://creativecommons.org/ licenses/by/4.0/).
TheFrench Society ofAnaesthesiaand Critical Care (Socie´te´ franc¸aised’anesthe´sie etde re´animation,SFAR) and the French AssociationfortheStudyoftheLiver(Associationfranc¸aisepour l’e´tude du foie, AFEF) have joined forces to generate original guidelinesdedicatedtothemanagementofliverfailureincritical caresetting.Theseguidelinesareintendedforawideaudienceof intensivists.Theaimisnottodetailthespecificmeasuresadopted in specialised or dedicated liverICU to manage thesepatients, but rather to specify the first-line diagnostic and therapeutic managementthatshouldbeinitiatedinanycriticalcareunit.The recommendations also focus on specifying the circumstances whereexpertadvicefromdedicatedcentres,mostoftenbackedby livertransplantunits,isrequired.TwentyFrench-speakingexperts werethus selectedby an organising committee that had been appointedbytheguidelinecommitteesapprovedbytheexecutive boardsofthetwoscientificsocieties.Theseexpertswereaskedto produceguidelinescoveringtwospecificareas:acuteliverfailure and cirrhoticdisease. Twobibliographicexperts also used pre-definedkeywordstoreviewalltheliteratureinthefieldpublished duringthepasttwentyyears.
2. Aimoftheguidelines
Theaimof theseguidelinesis toprovidea decision-making framework for physicians practicing in a general critical care settingwhoneedtotreatpatientswithsuspectedliverfailure.The grouptriedtoproduceaminimumnumberofrecommendations soastohighlighttheimportantfeaturestoberetainedinthetwo predefinedfields.Insituationsofdoubt,theweightoftheliterature wasconsideredtobemoreimportantthantheexperts’opinions. Thebasic rules of universal good medical practicein intensive carewereconsideredtobeknownandwereexcludedfromthe recommendations.The target audience is large, focusing on all professionalintensivists withtheexceptionofthoseworkingin structures dedicated to liver diseases, often backed by liver transplantcentres.
3. Definitions
Twofieldsweredefined:
AcuteLiverFailure(ALF)correspondingtoasudden deteriora-tioninliverfunctionduetotheappearanceoftoxicorinfectious agentsintheabsenceofunderlyingliverdisease;
liverfailureinpatientswithunderlyingchronicliverdiseaseor cirrhosis, frequently associated with Acute on Chronic Liver Failure(ACLF).
4. Methods
Theseguidelinesweredrawnupbyagroupofexpertsactingon behalfoftheAssociationfranc¸aisepourl’e´tudedufoie(AFEF)and theSocie´te´ franc¸aise d’anesthe´sieetdere´animation(SFAR).The organisingcommitteedefinedalistofquestionstobeaddressed anddesignatedexpertstoberesponsibleforeachquestion.The questionswereformulatedusingthePatientIntervention Com-parisonOutcome(PICO)model.
The Grade Method (Grade of Recommendation Assessment, Developmentand Evaluation)wasusedtocompilethese guide-lines. Following a quantitative analysis of the literature, the methodcanbeusedtoseparatelydetermine thequalityof the evidenceavailable,i.e.toestimatethelevelofconfidencerequired toanalysetheeffectsofthequantitativeintervention,andthelevel ofrecommendation.Thequalityofevidenceisratedasfollows:
highquality ofevidence: further researchis veryunlikely to affectconfidenceintheestimateoftheeffect;
moderatequalityofevidence:furtherresearchislikelytohave animpactonconfidenceintheestimateoftheeffectandcould changethisestimateoftheeffect;
lowqualityofevidence:furtherresearchisverylikelytohavean impactonconfidenceintheestimateoftheeffectandislikelyto changethisestimateoftheeffect;
verylowqualityofevidence:anyestimateoftheeffectisvery unlikely.
The level of recommendation is binary (either positive or negative)andstrongorweak:
strongrecommendation:werecommend(grade1+)ordonot recommend(grade1 )thisaction;
weak recommendation: we suggest (grade 2+) or do not suggest(grade2 )thisaction.
Thestrengthoftherecommendationswasdetermined accord-ingtokeyfactorsandvalidatedbytheexpertsafteravoteusingthe GradeGridmethod[1].
The compilationofa guideline requiredthat at least50% of voting participants had an opinion and that fewer than 20% ofparticipantsvotedfortheoppositeproposal.Thecompilationof a strongagreementrequiredtheapprovalofatleast70%ofthe votingparticipants.
5. Results
Nineteen experts and four coordinators agreed to address questionsconcerningthetreatmentofliverfailureanditspractical implementation in the general intensive care setting in two specificcontexts:
acute liver failure (ALF) defined as the development of hepatocellulardysfunctioninpatientswithoutpreexistingliver disease;
liverfailureinpatientswithunderlyingchronicliverdiseaseor cirrhosis, frequently associated with acute on chronic liver failure(ACLF).
ACLF is defined as a clinical syndrome of sudden hepatic decompensationobservedinpatients withchronicliverdisease andassociatedwiththefailureofoneormoreextrahepaticorgans. The experts summarised the work and applied the Grade methods, which resulted in 18 recommendations and three management algorithms. Six of these recommendations were strong (grade 1),six wereweak (grade 2)and six wereexpert opinions.Norecommendationwasformulatedinresponsetoone question.Aftertworoundsofscoringandvariousamendments,a strongagreementfromallvotingparticipantswasobtainedforall recommendations(100%).
5.1. Firstarea.Acuteliverfailure
Acuteliverfailure(ALF)isararedisease(fewerthan10cases permillionpersonsperyearinthedevelopedworld),characterised byrapidlyprogressiveliverdysfunctionassociatedwithafallin prothrombintime(PT)ratiolevelsobservedinpatientswithout anypreexistingliverdiseaseandinlessthan26weeks.
SevereALFdefinesasyndromecharacterisedbyaPTratioless than50%.
SeriousALFdefinesasyndromecharacterisedbyaPTratioless than50%incombinationwithencephalopathy.
There arealsoseveraldefinitions basedonthetime elapsing between jaundice and the development of encephalopathy. Fulminanthepatitis referstothe occurrenceof encephalopathy within15daysoftheonsetofjaundice,althoughtheoriginalterm ‘‘fulminanthepatitis’’isclassicallyusedtodescribesevereALF.The terms‘‘severe’’or‘‘serious’’ALFareusedintheremainderofthis text.TheprognosisofALFhasgraduallyimprovedinrecentyears, witha 2-year survivalrateofabout 90%in theeventof trans-plantationandof90%aftersevereALFduetoparacetamolwithout transplantation[2].
Analgorithm (Fig.1)and atable(Table1)summarisingALF managementareproposed.
6. Question1:Inpatientswithacuteliverfailure,which etiologicalexamsshouldbeperformedtoreducemorbidityand mortality?
R1–Inpatientswithsevereacuteliverfailure,werecommend thedeterminationofserumacetaminophenlevels,serologyfor HepatitisA(IgMVHA)andHepatitisB(HBsAgandanti-HBcIgM) viruses,urinarytoxins(amphetamine,cocaine),andthe perfor-manceofanechocardiographyandhepaticecho-Doppler.
(GRADE1+),STRONGAGREEMENT
Fig.1.Proposedalgorithmforthemanagementofacuteliverfailure.
Table1
Managementofextrahepaticorganfailureinpatientswithacuteliverfailure.
Whattodo Whatnottodo
Centralnervoussystem Encephalopathyshouldbemonitoredfrequently
Maintainserumsodiumlevelsbetween140and145mmol/L
Monitoringofbloodglucosewillprobablyberequiredatleastevery2hours TrachealintubationandsedationintheeventofprogressiveHE(Glasgow<8) Practicestominimisethedepthofsedationarerecommended
TranscranialDopplerultrasound ICP:nospecifictreatment
Administrationofsedativessuchas benzodiazepinesandpsychotropicdrugs(such asmetoclopramide)
Useoftreatments(lactulose,rifaximin)to lowerammonialevels
Respiratorysystem Standardlungprotectiveventilatorstrategy(accordingtospecific recommendations)
Cardiovascularsystem Assessmentofvolumestatus,cardiacoutputandcardiacfunction(rightand left-sidedfunction)
Fluidexpansionusingcrystalloidfluidsasfirstchoice Norepinephrineinfusionforrefractoryhypotension
Renalsystem Renalreplacementtherapyaccordingtospecificrecommendations Useofnephrotoxicdrugs,including non-steroidalanti-inflammatorydrugs Gastrointestinalsystem Stressulcerprophylaxisaccordingtospecificrecommendations
Coagulation Routinecorrectionofcoagulation:restrict clottingfactorsadministrationunlessactive bleeding
Immunesystem Empiricalbroadspectrumantibioticsshouldbeadministeredtopatientswith worseningHEorsignsofSIRS
6.1. Rationale
Among the most common causes of ALF, acetaminophen-inducedhepatotoxicity(whetherintentionalornot)isthemost commoncauseofsevereALFandanindicationforemergencyliver transplantation(22%inFrance)[3,4].ALFmayalsobecausedby HepatitisAandBinfections(14.6%),otherdrugs(antimicrobialand antiepilepticagents,statins)(9.4%) orothervarioustoxicagents (herbalsupplements,cocaine,ecstasyortheingestionof mush-rooms)(4.8%).
Lessfrequently,otherscausesofALFmaybeidentifiedsuchas autoimmunehepatitis, Wilson’s disease, metabolic disease, etc. Theseaccountfor28.8%ofpatientsonthelivertransplantwaiting list.Noteworthy, thecauseof ALFremains unknownin 25% of patients,despiteintensiveinvestigation.Clinicalassessmentand extensive search in patient’s medical history are important elementstodeterminethecauseofALF.
Early recognition of the reason for ALF is crucial to guide potential specific management and predict its outcome
[3].The transplant-freesurvivalofpatients withALFrelated to acetaminophen,hepatitisA,hypoxichepatitisorpregnancyis50%. Conversely,whenALFisassociatedwithhepatitisB,drugsother thanacetaminophen,autoimmunehepatitis,Wilson’sdiseaseor Budd–Chiarisyndrome,thetransplant-freesurvivalrateislower than25%.
AbdominalDopplerultrasoundshouldbeperformedrapidlyto excludechronicliverdiseasesuggestedbyascites,hepatomegaly, dysmorphicliverormalignantinfiltrationoftheliver,andtoverify thepermeabilityofvessels(hepaticveins,portalvein). Echocardi-ographyshouldalsobeperformedwhenacuteischemic hepato-cellular injury is suspected [critically ill patients withprimary cardiac or circulatory failure, elderly individuals, underlying historyofheartdiseaseorarrhythmia,patientswithacuterenal failure, serum level of aspartate transaminase (AST) exceeding thoseofalanineaminotransferase(ALT)][5].
7. Question2:Inpatientswithacuteliverfailure,whichspecific treatmentsshouldbeinitiatedrapidlytoreducemorbidityand mortality?
R2.1 – In patients with acetaminophen-induced acute liver failure,werecommendtheinitiationofN-acetylcysteine ther-apywithoutwaitingfor,andregardlessof,theresultsofserum acetaminophendeterminations.
(GRADE1+)STRONGAGREEMENT
R2.2–Inpatientswithacuteliverfailurewhateverthe aetio-logy,wesuggesttheinitiationofN-acetylcysteinetherapyto improvemorbidityandmortality.
(GRADE2+)STRONGAGREEMENT
7.1. Rationale
In a retrospective analysis of patients with acetaminophen-induced liver failure, fewerpatients in the group treated with acetylcysteinecomparedtonon-treatedgroupprogressedtograde III–IVencephalopathy(21/41[51%]versus43/57[75%],P<0.05) or died (15/41 [37%] treated versus 26/41 [63%] non-treated, P<0.05) [6]. A placebo-controlledrandomised trial in patients with acetaminophen-induced liver failure demonstrated an increase in the 21-day survival rate of those treated with acetylcysteine (12/25 [48%] versus 5/25 [25%] respectively,
P=0.037) [7]. Additionally, fewer treated patients developed clinicalsignsofcerebraloedema(40%versus68%,P=0.047)and neededvasoconstrictorstomaintainblood pressure(48%versus 80%;P=0.018)[7].Arecentmeta-analysisfoundthat acetylcys-teinewassignificantlysuperiortoplacebo,resultinginlowerlevels ofhepatotoxicity(18%versus58%[RR0.31,95%CI0.26–0.39])and mortality(0.7%versus6%[RR=0.12,95%CI0.04–0.38])[8].
Ithasbeensuggestedthatthetherapeuticuseofacetylcysteine mayalsobenefitpatientswithnon-acetaminophen-relatedALF.A recent meta-analysis analysed the results of four prospective clinicaltrialsregardingthesafetyandefficacyofacetylcysteinein patients (both adults and children) with ALF not related to acetaminophen poisoning [9]. A totalof 331 patients received treatment with acetylcysteine as compared to 285 patients in controlgroup[10–12].Nostatisticaldifferencewasseenbetween the two groups in terms of overall mortality. However, liver transplant-freesurvival(41%versus30%,OR=1.61,95%CI1.11– 2.34,P=0.01)andpost-transplantsurvival(85.7% versus71.4%, OR=2.44,95%CI1.11–5.37,P=0.03)weresignificantlygreaterin treated patients. We performed a newmeta-analysis including adultpatientsonlyandtworecentstudies[13,14][10,15],which displayed improvements in overall survival (four studies with 499 patients: 76% versus 59%, OR=2.30, 95% CI 1.54–3.45 P<0.0001)andlivertransplant-freesurvival(threestudieswith 419 patients: 64% versus 26%, OR=4.81, 95% CI 3.22–7.18, P<0.0001).However, becauseof methodologicalbias,we have producedalowergrade(GRADE2)recommendationfortheuseof acetylcysteineinnon-acetaminophen-relatedALF.
During these studies, acetylcysteine was predominantly administeredintravenously,atvaryingdoses.Theadverseeffects identified included nausea, vomiting, diarrhea or constipation. Acetylcysteinecouldcauseskinrash(<5%)ortransient broncho-spasm(1–2%).
In theprospectivedouble-blindtrialperformed byLeeet al.
[10], the beneficial effects of acetylcysteine on transplant-free survival wereconfined to patients with gradesI–II coma. This suggeststhattheintervalbetweendrugingestionandtreatment with acetylcysteine is closely related to the outcome, so that acetylcysteine shouldbestarted asearlyaspossible. Acetylcys-teine hascomplex antioxidant and immunologiceffects whose mechanismsofactionarenotcompletelyunderstood.
R2.3–Inpatientswithacuteliverfailure,whateverthe aetio-logy,theexpertssuggestthatadviceshouldbeobtainedfroma livertransplantationcentreinordertodiscuss:
1.Second-line aetiological investigationsif theresults of first-lineexaminations(seeR1)arenegative
2.Anindicationforlivertransplantation EXPERTOPINION,STRONGAGREEMENT
7.2. Rationale
7.2.1. Secondaryetiologies
In France, other causes of ALF identified after second-line investigations(autoimmunehepatitis,Wilson’sdisease,metabolic disease,etc.)accountfor28.8%ofthecasesonthelivertransplant waiting list for ALF. If the initialinvestigation is negative, the followinginvestigationsshouldbecarriedout:
Antinuclearantibodies,antimitochondrialantibodies,anti-LKM antibodies, smooth muscle antibodies (autoimmune hepatitis); anti-HEVIgM(hepatitisEvirus);Anti-HSVIgM (herpessimplex virustypes1and2);serumandurinarycopper,serum caerulo-plasmin,(Wilson’sdisease).PregnancyrelatedALF:therearetwo hepaticemergencieswhichcanoccurduringthethirdtrimesterof pregnancy:HELLPsyndromeandacutefattyliverofpregnancy.
7.2.2. Livertransplantation
Severe ALF defines a syndrome characterised by a PT ratio<50%.Thereshouldbeacontactwithalivertransplantation centreforeachpatientwithsevereALFallowingdiscussionabout diagnosticandconsiderationfortransferinadedicatedcentre.
Hepaticencephalopathy(HE)isofkeyprognosticimportancein ALF.Neurologicalsymptomsareassociatedwithoverallsurvival (90.1%inALFpatients withoutHE versus37.8%in ALFpatients withHE;P<0.0001)[16].HEgradesareassociatedwithoutcome. Short-termtransplant-freesurvival(threeweeks)hasbeenshown tovaryconsiderably,from52%inpatientswithgrade1–2HEto 33%inthosewithgrade3–4HE[3].Thedevelopmentofgrade3–4 HEisassociatedwithbrainoedemaandintracranialhypertension in38%to81%ofpatients[17,18].HEisthekeyindicatorforLT. ThiscriterionisusedtoselectcandidatesforLT(King’sCollege criteria in ALF patients related or not to paracetamol; Clichy– Villejuifcriteria)[4,19].Thesecriteriaareprognosticindicatorsof transplant-freesurvivalandareusedintheselectionofpatientsfor LT.Thehistorical criteriaarethosedeveloped byKing’sCollege Hospital:inacetaminophen-inducedliverfailureapoorprognosis iscorrelatedwithapHlowerthan7.3,prothrombintimelonger than100s,acreatininelevelhigherthan300mol/landHEmore severe than grade 3. Furthermore, blood lactate levels higher than 3.5mmol/L after4hours, or 3.0mmol/L after 12hours of managementand theearlyrestoration ofintravascularvolume, areindicativeofapoorprognosis.Recentstudieshaveconfirmed theclinicallyacceptablespecificitybutmorelimitedsensitivityof thesesystems (50% to60%) [20–23].In orderto addressthese limitations,awidevarietyofalternativeprognosticsystemsand markershavebeenproposed toreplaceorsupplement existing criteria(forexample,to identifya patient fortransferto anLT centre).TheseincludetheuseoffactorVlevelslessthan20%[4], orthecombinationofcriteria:vasopressorsneedinacontextof organfailure [20,24], the routine laboratorymeasurements in-cludingthosesuchasammonia(withathresholdof100
m
mol/L)[17,18,25,26]orbilirubin(thresholdrangingfrom140
m
mol/Lto200
m
mol/L[4,27]),orcompositelaboratorydeterminationssuch astheMELDscore[22,27,28].Thereisacloserelationshipbetweenelevatedarterialammonia levels and the development of encephalopathy, there being a greaterriskofintracranialhypertensionwhenammonialevelsare sustainedbetween150and200
m
mol/L[17,18,26].Alltheseclinicaland biological criteriaarecorrelated tothe severityofliverfailurebuttheyneverthelesshaveanindependent prognosticvalue[4,17,22,26–28].
8. Question3:Inpatientswithacuteliverfailure,which symptomatictreatmentshouldbeinitiatedtoreducemorbidity andmortality?
R3–Inordertoreducemorbidityandmortalityinpatientswith acuteliverfailure,theexpertssuggestthatextrahepaticorgan failure shouldbetreatedearlyand anyaggravatingfactors shouldbeprevented,asshownintheTablebelow.
EXPERTOPINION,STRONGAGREEMENT
8.1. Rationale
PatientswithALFdevelophypotensionwithsystemic vasodi-lation and volume depletion that reflect the severity of the underlyingliverfailure[29].MostpatientswithALFwillhavea hyperdynamiccirculation. Haemodynamicassessment is neces-sarytodeterminevolumedepletionandbothrightandleft-sided
cardiacfunction,becausesomepatientscandeveloprightorleft cardiacdysfunction[1].Thereisnospecificliteraturetoguidethe typeoffluidstouseorthechoiceofvasopressoragents[30,31].
Thereisanevidenceofadrenaldysfunctionin50%ofpatients withALF,whichisatleastrelative[32,33].Onlyoneretrospective studyincluding40patients,reportedthattheuseof hydrocorti-sone(300mg perday) could reducethe needfor vasopressors
[34]. In the ICU population, the diagnosis and management of critical illness-related corticosteroid insufficiency remains a matterofdebate[35–37].
Trachealintubationis usuallyindicatedwhenGlasgowcoma score is less than 8. Mechanical ventilation settings should be protective,asstipulatedinguidelinesfromspecialistcriticalcare societies [38]. There havenot beenanyspecific studieson ALF populations[38,39].HighlevelsofPEEP(>10cmH2O)couldbe
associatedwithapotentialriskofhepaticcongestion[40–42]. Norandomisedcontrolledtrials(RCT)haveevaluatedsedation practicesinpatientswithALF.IntheICUpopulation,studieshave shownthatprotocol-basedsedation(tominimiseitsdepthand duration)seemstoreduceoverallmorbidityandmortality.The useof benzodiazepines should be avoided [43,44]. Inpatients withacuteorchronicencephalopathy,ameta-analysis(8RCTs; n=736patients)showedthatflumazenilloweredthe encepha-lopathyscore,alsosuggestinga deleteriouseffectof benzodia-zepinesinthispopulation[45].Dexmedetomidineshouldbeused withcaution,asits metabolismis exclusivelyhepatic[46].No studieshaverecommendedornotthesystematicuseofsedation in patients with an altered Glasgow Coma score without intracranialhypertension(ICH),aclassicallydescribed complica-tion of HE that affects 20% of patients with ALF[16]. Regular monitoring must be ensured of patients with high-grade encephalopathy(grades3and4).Twomulticentreobservational studiesinpatientswithsevereencephalopathydidnot demon-strateasignificantdifferencein1-monthmortality,whetherthe patientsweremonitoredusingICPdevicesornot(pooleddata: RR=0.79, 95% CI 0.61–1.02) [47,48]; ICP devices have been associatedwithhaemorrhagiccomplications(7%to20%ofcasesin patientswithALF)[47–49].TranscranialDopplerultrasoundisa usefulmonitoringtoolthatcouldbeusedfirst-lineinthiscontext. Therearenospecifictherapeuticoptionsforraisedintracranial pressureinpatientswithALF[50–55].
8.1.1. Coagulation
A recent multicentre observational study including 1770patientswithALFreportedbleedingcomplicationsinonly 187 patients (10%), which included 22 (1.5%) post-procedural bleeding episodes [56]. Eighty-four per cent of spontaneous bleedingepisodesoriginatedfromtheuppergastrointestinaltract. Anin-depthanalysisofcoagulationabnormalities (thrombocyto-peniaandprolongationoftheINR)inALFpatientssuggestedthat most patients had rebalanced haemostasis between pro- and anticoagulantfactors[57].Prophylacticadministrationof coagu-lationfactorsprecludestheassessmentofthenaturalevolutionof thedisease.Thereisnosupportfortheuseofcoagulationfactors and this should be limited to active bleeding or invasive procedureswithahighriskofcomplications.
8.1.2. Renalreplacementtherapy
No RCT have been performed specifically to evaluate the strategy and timing of its initiation in patients withALF [16]. Regionalcitrateanticoagulationshouldbemonitoredbecauseof itspotentialmetaboliceffectsinpatientswithALF[58].
8.1.3. Liversupportdevices
Twowell-designedRCTincluding115patientswithALF(not related tohypoxichepatitis)failedtodemonstratea significant
reductioninmortality(pooleddata:RR=0.82;95%CI0.42–1.59)
[59,60].Theplaceofliversupportsystemsinthemanagementof
patients withALF needtobebetterdefined.In anycase, these techniquesshould not delay transfer to a liver transplantation centre.
ALFisfrequentlyassociatedwithelectrolyteandmetabolic disturbances,particularly in patients with hyperacute ALF or whenitisassociatedwithacutekidneyinjury[61]. Hypoglycae-miaisawell-knowncomplicationofsevereALF[2,61].Itsclinical featurescanbeconfusedwiththoseofhepaticencephalopathy. Blood glucose parameters should therefore be monitored at leastevery2hours.Nostudieshaveevaluatedtheoptimaltarget forbloodglucoselevels.Hyponatremia(sodium<130mmol/L) is common in patients with ALF [62]. There is a correlation betweenhyponatremia and intracranial pressure.It has been shownthattheinfusionofhypertonicsalinetomaintainserum sodium levels at between 145 and 155mmol/L significantly decreased ICP [54]. However, a serum sodium level above 150mmol/Lisdeleteriousand shouldbeavoided.Inpractice, werecommendtargetingthesodiumlevelatbetween140and 145mmol/L.Anycorrectionsshouldnotexceed10mmol/Lper 24h[63]. Electrolyte disturbances,such as serum phosphate, are commonly observed in patients with ALF, and should be monitoredandcorrected.
Patients with ALF and organ failure have increased energy expenditure similartothat of otherICU patients. There are no specificnutritionalguidelinesinpatientswithALF.
The use of osmotic laxatives (lactulose) or non-absorbable antibiotics(rifaximin)tolowerammonialevelsisnot recommen-ded[61].
Although there are no data to support its use, stress ulcer prophylaxisisusuallyrecommendedinthisat-riskpopulation[61]. PatientswithALFhaveincreasedsusceptibilitytoinfections. Bacterialinfectionshavebeendocumentedin60%–80%ofpatients withALF, andfungal infectionsoccur in one third of patients
[64].Empiricalbroad-spectrum antibioticsshould be adminis-trated to ALF patients if there are signs of sepsis and/or of worsening encephalopathy [65]. These broad-spectrum anti-bioticsshouldcovercommonorganismssuchasenterobacteria, staphylococcalorstreptococcalspecies,orasa functionofthe ecologyintheunit[64].
8.2. Secondarea.CirrhoticICUpatientsandacute-on-chronicliver failure
Cirrhosis is an end-stage form of chronic liver disease.The prevalence of cirrhosis is estimated at between 2000 and 3300 people per million inhabitants. There are approximately 700,000 patients withcirrhosis in France, 10,000 to 15,000 of whomdieeveryyear.Awiderangeofdiseasesandconditionscan damagetheliverandleadtocirrhosis.Ninetypercentofcirrhosis casesarerelatedtochronicalcoholabuse,chronicviralhepatitis (hepatitisBandC)andfattyliverdiseaseinacontextofmetabolic syndrome.In terms ofitspathophysiology, cirrhosisconsistsin thegradualreplacementofhealthylivertissuebyannularfibrosis that is responsible for destroying the architecture of the liver parenchymaand blockingintrahepaticportal blood flow.These architecturalchangesareassociatedwithlife-threatening compli-cationsrelatedtoportalhypertensionand/orhepaticdysfunction that may requireICU hospitalisation. These complications may beeitherspecifictoliverdisease(ascites,spontaneousbacterial peritonitis, variceal haemorrhage, hepatic encephalopathy or hepatorenalsyndrome) or non-specific (infection, acute kidney failure). In addition, given the prevalence of this disease, it is frequent that the reason for initial ICU admission is not a complicationof cirrhosis but ratheranother condition likely to
lead to organ failure such as pneumonia-related septic shock, community-acquiredperitonitis,stroke,trauma,etc.Underthese circumstances,evenifcirrhosiswascompensatedorevenoftennot diagnosed at admission, the precipitating event is likely to decompensatetheliverdisease.
CirrhoticpatientsrequiringICUhospitalisationhaveahighrate ofin-hospitalmortality(rangingfrom30%to50%).Theseverityof cirrhosiswasinitiallyestablishedbytheChild–Turcottescore[66]
subsequentlymodifiedbyPugh[67].AnewentitynamedACLFhas recentlybeendescribed [68]as acombination ofacute decom-pensationofcirrhosis,oneorseveralextrahepaticorganfailures andhighshort-termmortality[68].ACLFisadynamicsyndrome andcanoccurinpatientswithorwithoutapriorhistoryofacute decompensation.ThediagnosisofACLFanditsclassificationinfour gradesofseveritybasedonthenumberoforgansfailingismade using theCLIFSOFAscore Supplementary Tables 1and 2, [68– 70].ItshouldalsobenotedthatanACLFgrade basedonorgan failure hasbeen shown to better predict the outcomesof ICU cirrhoticpatientsthanstandard prognosticmethodssuchasthe MELDorChild–Pughscores.
Theseguidelineswillfocusonthemanagementofcirrhosisand its complications requiring ICU hospitalisation, and provide recommendations based on the best evidence available in the literature.
9. Question4:Inpatientswithcirrhosis,whichcriteriashould beusedtoguideadmissionornon-admissiontoanICUto improvetheirprognosis?
R4–Wedonotsuggestdenyingtheadmissionofpatientswith cirrhosistotheICUsolelybecauseoftheirunderlyingcirrhotic condition.
(GRADE2 )STRONGAGREEMENT
9.1. Rationale
The proportion of cirrhotic patients admitted to ICUs is increasing[71–74].However,todate,therehavebeennoobjective and specificcriteriatoguidetheadmission ofthesepatientsto criticalcare.
Therecentdefinitionofanewclinicalentity,Acute-on-Chronic Liver Failure (ACLF), as an acute decompensation of cirrhosis associatedwithorganfailure(s)definedusingtheCLIF-SOFAscore couldbeusefultoensuretheearlyidentificationofpatientswho couldbenefitfromICUadmission.Indeed,ACLFisassociatedwitha 28-daymortalityrateofatleast15%[68].
AmongtheelementsthatcanguideadecisionofICUadmission, initialseverityscoreshavebeenanalysedinseveralreviewsofthe literature[75–79]includingobservationalstudiesthatevaluated thesescores[69,70,80–99],aswellasthemostrecentones[100– 108].Topredictshort-termmortality(intheICUorinhospital), standard organ failure scores such as SOFA (23 studies, AUC: 0.70to0.95),CLIF-SOFA(7studies,AUC:0.72to0.83)orCLIF-C Acute on Chronic Liver Failure (1 study, C-index0.76) display similar performance [99,100,106]. Their performance is better thanthatofgeneralICUscoressuchasAPACHE2(20studies,AUC: 0.66to0.90),APACHE3(5studies,AUC: 0.72to0.91)or SAPS2 (6studies,AUC:0.74to0.89).Studiesreportingtheprognosisat 6months have foundsimilarresults[109]. Theperformance of liver-specificscoresislower,withMELD(18studies,AUC:0.70to 0.93) being slightly better than Child-Pugh (21 studies, AUC: 0.55to0.87)[110].ScoresderivedfromMELD(MELD-Na,iMELD, MESO) do not have a better prognostic value than MELD
ofdecompensationofcirrhosis,isnotassociatedwithprognosisin
theICU[69,111].
ThenumberoforganfailuresatadmissiontotheICUisalso associatedwithshort-termmortality[69,70,74,77,82,112,113].A recentandlargemulticentrestudyconductedon17,044cirrhotic patientsinintensivecarereportedahospitalmortalityratelower than 50% in patients with three or more organ failures, thus delegitimisingdecisionsnottoadmitacirrhoticpatienttotheICU basedonliverdiseasealone[113].
In addition, the reason for ICU admission also influences prognosis,withdigestivehaemorrhageclassicallybeingassociated with a better prognosis and septic shock with a poorer one
[68,70,105,114].Finally,amongcirrhoticpatientsadmittedtoan
ICU, those who have been directly admitted have a better prognosis than those who were initially admitted to the HepatologyUnit,thushighlightingthevalueofearlyadmission
toICU[68,69,72,115].
As with non-cirrhotic patients, the prognosis for cirrhotic patientsatICUadmissionisthereforeverylargelyassociatedwith the presence of organ failure(s) graded using different scores. However,noneofthesescoreshassufficientpredictivepowerto guideindividualpatientprognosis.Thereisthereforenoreliable indicator of the futility of resuscitation in cirrhotic patients, regardlessoftheseverityofcirrhosisornumberoforganfailures
[116].
Finally, several studies have shown that the ICU prognosis of cirrhotic patients has improved in recent years (period effect independently associated with mortality),
[72,89,100,108,113,114,117,118]whether thereason for
admis-sionwas variceal bleeding [119] or septic shock [74,120]. The maintenance of life-sustaining therapy during the ICU stay is consideredinquestion10.
10. Question5:InpatientswithcirrhosishospitalisedinanICU, whichspecificmeasurestomanageacutekidneyinjuryshould beusedtoreducemorbidityandmortality?
R5.1–Todefineandassesstheseverityofacutekidneyinjury (AKI)inpatientswithcirrhosis,theexpertssuggest:
1.UsingthemodifiedKDIGOcriteriaspecificallyrefinedby theInternationalClubofAscites
2.ManagingAKIaccordingtoitsseverityandtothe algo-rithmproposed(Fig.2)
3.Notcontraindicatingrenalreplacementtherapyinthese patientsjustbecauseoftheirunderlyingcirrhoticcondition.
EXPERTOPINION,STRONGAGREEMENT
10.1. Rationale
In non-cirrhotic patients, the reference for diagnosing and assessingtheseverityofacutekidneyinjury(AKI)is theKDIGO classification [121,122]. The limitation of this classification in patients with cirrhosis is the use of urinary flow. Indeed, in advanced stages, cirrhosis isoften associated witholiguriaand waterandsaltretention,oedemaandascites,whilerenalfunction remainsnormal.Conversely,theuseofdiureticstotreatascites and oedemacan leadtoan abnormallyhighurine outputwith impairedrenalfunction.ExpertsintheInternationalAscitesClub havethereforeproposedusingtheKDIGOclassificationtodiagnose andassesstheseverityofAKI,basedonthesamecriteriaasfor serum creatinine without the urine output criterion (Table 2)
[122].Severalstudieshaveshownthatthismodifiedclassification
has a good prognostic value with early mortality being well
correlatedwiththeseverityofAKI[123–127].Itisimportantto notethatthedefinitionisbasedonthedynamicchangeinserum creatininefrombaseline andno longeron a fixedthresholdfor serum creatinine.Indeed,as a marker of renal function,serum creatinine has many limitations in patients with cirrhosis, including:
adecreaseinmusclecreatine,aprecursorofcreatinine,related tofrequentsarcopeniainthesepatients[128];
an increasein thedistribution volumerelated toascites and oedema;
interferencewithcreatininemeasurementtechniquesrelatedto elevatedserumbilirubinlevels[129,130].
UseofthemodifiedKDIGOclassification(Table2)isimportant becauseitenables themanagement ofacuterenal failuretobe adapted as a function of its severity(Fig. 2). So, although the evictionofriskfactorsmaybesufficientatanearlystage,albumin fluidtherapyisnecessaryintheeventofprogressiontostages2or
3[121].Finally, althoughtheprognosiswithrenal replacement
therapy is very poor in cirrhotic patients with AKI, it is not contraindicatedprovidedthatitsdurationisplannedtobeshort and that it is integrated in a therapeutic plan, such as liver transplantation,orinacontextofreversibleprecipitatingevents, suchassepsis(Table3)[131].
R5.2–InpatientswithcirrhosishospitalisedintheICU,we suggesttreatinghepatorenalsyndromewithavasoconstrictor agent (terlipressin as first-line therapy) and concentrated albumin
GRADE2+,STRONGAGREEMENT
10.2. Rationale
Severalmeta-analyseshavesuggestedabeneficialeffectofthe combination of a vasoconstrictor and albumin on short-term survival and hepatorenal syndrome (HRS, defined in Table 2) regression[131–133].
Theobjectiveofvasoconstrictionistocounterbalance splanch-nicarterialvasodilationinordertoimproverenalperfusion.The referencetreatmentisterlipressin.ACochranemeta-analysisthat included nine randomised studies (394 patients) comparing differentvasoconstrictorsincombinationwithalbuminconcluded thatterlipressinwasthemosteffectiveintreatingHRS[134].Most recentstudieshavesuggestedthattheresponserateintermsof animprovementinrenalfunction(completeorpartialresponse) ranges from 64% to 76% [134]. However, these rates are significantlylowerintheeventofrecurrentHRS(20%ofcases). Terlipressin is usually administered intravenously (bolus injec-tions)ataninitialdoseof0.5mgto1mgevery4to6hours.The dosecanthenbegraduallyincreasedtoamaximumof2mgevery 4to6hoursiftheserumcreatinineleveldoesnotfallbymorethan 25%.Treatmentshouldbemaintaineduntilacompleteresponse isobtainedorforamaximumof14daysinthecaseofapartial response. The continuous administration of terlipressin at the same daily doses (2 to 12mg/24h IVSE) appears to be an interesting alternative asit remains as effectiveas bolusesbut enables lowerdaily doses,therebyreducing theriskof adverse eventssuchascardiacorintestinalischemia,pulmonaryoedema, or distal necrosis [135]. More recently, another meta-analysis pointed outthat norepinephrinemightbea reliablealternative in patients with a central venous catheter (terlipressin can be administeredviaaperipheralvein)[133].Indeed,nodifferencein the reversibility or relapse of HRS was found between the terlipressin+albumin and norepinephrine+albumin arms, al-thoughthenumbersofpatientsincludedinthesestudiesremain small.Conversely,thecombinationofmidodrineandoctreotideis lesseffectiveandshouldnotbeused[135,136].
The addition of albumin can be discussed insofar as its beneficial effects haveonly been demonstrated in combination withvasoconstrictors.Indeed,whiletheadministrationofalbumin might improvesystemic haemodynamics by increasing cardiac output and through its antioxidant and anti-inflammatory properties[137],nostudyhasevercomparedastrategycombining vasoconstrictors and albumin with another that only used vasoconstrictors. Therefore,thedosesusually recommendedfor albumin(1g/kgbeforetheinitiationofvasoconstrictortreatment
Table2
ModifiedKDIGOcriteriafordefinitionofacutekidneyinjuryinpatientswithcirrhosis. Baselineserumcreatinine
Avalueofserumcreatinineobtainedinthepreviousthreemonths,whenavailable,canbeusedasbaselineserumcreatinine
Inpatientswithmorethanonevaluewithintheprevious3months,thevalueclosesttotheadmissiontimetothehospitalshouldbeused Inpatientswithoutapreviousserumcreatininevalue,thevalueonadmissionshouldbeusedasbaseline
Definitionofacutekidneyinjury
Increaseinserumcreatinine26.5mmol/Lwithin48hours or
Apercentageincreaseserumcreatinine50%frombaseline,whichisknownorpresumedtohaveoccurredwithintheprior7days Stagingofacutekidneyinjury
Stage1:increaseinserumcreatinine26.5mmol/Loranincreaseserumcreatinine1.5-foldto2-foldfrombaselinevalue Stage2:increaseserumcreatinine2-foldto3-foldfrombaselinevalue
Stage3:increaseserumcreatinine3-foldfrombaselinevalueorserumcreatinine353.6mmol/Lwithanacuteincrease26.5mmol/Lorinitiationofrenal replacementtherapy
Table3
Diagnosticcriteriaofhepatorenalsyndrome(allthecriteriamustberespectedtoretaintheHRS). Diagnosisofcirrhosisandascites
Diagnosisofacutekidneyinjurystage2or3accordingtoKDIGOcriteria
Noresponseafter2consecutivedaysofdiureticwithdrawalandplasmavolumeexpansionwithalbumin(1g/kgofbodyweight) Absenceofshock
Norecentuseofnephrotoxicdrugs(NSAIDs,aminoglycosides,iodinatedcontrast...) Nomacroscopicsignsofstructuralinjurydefinedas
Absenceofproteinuria(>500mg/day)
Absenceofmicrohaematuria(>50RBCsperhighpowerfield) Normalfindingsonrenalultrasonography
and then 20 to40g/day) [121] and the duration of treatment remainempirical, as do thehaemodynamic objectivesthat are notdefined.Uncertaintiesregardingthespecificeffectofalbumin, and recent changes to the definition of HRS may modify the interpretation of the results of previous studies and result in downgradingthelevelofevidenceoftherecommendation. 11.Question6:Inpatientswithcirrhosishospitalisedinthe ICU,whichspecificmanagementofsepsisshouldbeusedto reducemorbidityandmortality?
R6–Inordertoreducethemorbidityandmortalityofcritically illpatientswithcirrhosis,whateverthesymptomsandorgan failure(s),wesuggest:
1.Performingasystematicsearchforinfectionthatshould includemicrobiologicalandcytologicalexaminationofascites fluid(aconcentrationofpolymorphonuclearcells>250/mm3 intheasciticfluidwillconfirmthediagnosisofspontaneous bacterialperitonitis),
2.Initiatingearlyempiricalantibiotictherapythatshouldbe tailored tothesuspectedsiteoftheinfection,thecausative pathogenonceithasbeenidentified,andthelocalecology
GRADE2+,STRONGAGREEMENT
11.1. Rationale
Onethirdofcirrhoticpatientsdevelopaninfectionatadmission orduringhospitalisation,whichisfourtofivetimeshigherthan theincidenceinthegeneralpopulation.Infectionisan indepen-dentriskfactorformortality,accountingforonethirdtoonehalfof the causes of death in cirrhotic patients [138,139]. Several retrospective observational studies have shown an association betweendelayedantibiotictherapyandmortality[140–144].Ina studyincluding126cirrhoticpatientswithspontaneousbacterial peritonitis-relatedsepticshock,theadjustedORassociatedwith mortalitywas1.86(95%CI1.10–3.14)perhourofdelay(P=0.02)
[142].Another studyof 852cirrhoticpatientswithbacteraemia
showedthatadelayofmorethan24hoursbetweentheonsetof bacteraemiaand theinitiationof appropriateantibiotictherapy wasassociatedwithasix-foldincreaseinmortality[OR=6.06(90% CI4.28–8.58;P<0.0001)][143].Antimicrobialtreatmentshould thereforebestartedearlywhenthereisanyclinicalorbiological suspicionofsepsis.Worthyofnoteisaworseningof encephalopa-thyshouldsuggestaninfectionasitstriggeringfactor[145,146]. CRP and PCT serum levels were reported as being reliable biomarkersof infectionin cirrhotic patients in a meta-analysis that included10 studies(n=1144 patients) [147].However, it should be remembered that the diagnostic threshold for CRP decreaseswiththeseverityofcirrhosis(10mg/Lonaverageinmild cirrhotic patients, 5mg/L for Child C patients) [148]. Initial antibiotictherapyshouldtargetenterobacteriaandGram+Cocci, which account for the majority of causal pathogens in this populationinwhommorethanhalfofinfectionsarespontaneous bacterialperitonitisandurinarytractinfections[139].Thelocal microbiologicalecologyshouldalsobeconsideredwhenchoosing empirical antibiotic therapy. In a recent meta-analysis (eight studies,1074positive ascitescultures) [149],theaverage resis-tancerateto3rdgenerationcephalosporins(C3G)wasonethirdin community-acquiredSBPandtwothirdsinhealthcare-associated infections.
Concerning haemodynamics, cirrhotic patients are prone to develophypoxichepatitisthatcanworsenliverfailure.Thelatter maybemultifactorial[29]:hypovolaemiaandvasoplegia,aswell as right or left heart failure, pulmonary arterial hypertension
associatedwithportalhypertension,hepatopulmonarysyndrome andacutepulmonaryoedemaassociated withvolumeoverload. There isa lackofspecificdataon themanagement ofsepsisin cirrhotic patients, and the approach is currently based on guidelinesforthegeneralICUpopulation[150].Nospecificdata on the use of vasopressors are available outside observational studies[151].Ameta-analysis(16studies,1507patients)showed that halfof cirrhoticpatients withsepticshock,had, at leasta relative, corticosteroidinsufficiency,which wasassociatedwith higherratesof sepsisand hospitalmortality[152].An observa-tional study [153] and a double-blind RCT [154] included 150 cirrhoticpatients with septic shockin orderto assess the impact of 50mg hydrocortisone every 6h on mortality (very lowlevelofevidence).TherewasnosignificantdifferenceinICU mortality (pooleddata: RR=0.78; 95% CI 0.58–1.05). However, hydrocortisone was associated with a higher rate of shock re-covery(pooleddata:RR=1.49;95%CI1.17–1.92;P=0.001).The studywiththehighestlevelofevidence[154]reporteda signi-ficantlyhigherrateofgastrointestinalbleedinginthe hydrocorti-sonearm(RR=3.00;95%CI1.08–8.36;P=0.02).Overall,weare not able to produce a recommendation regarding the use of hydrocortisoneincirrhoticICUpatients;theirmanagementneeds tobeevaluatedonacase-by-casebasisinacontextofrefractory shock.
12. Question7:InpatientswithcirrhosishospitalisedinICU, whenconcentratedalbuminshouldbeadministeredtoreduce morbidityandmortality?
R7.1–IncirrhoticpatientshospitalisedintheICU,we recom-mendtheuseofaconcentratedalbumininfusionafter high-volumeparacentesis(morethan4to5litresofascitesfluid removed)
GRADE1+,STRONGAGREEMENT
12.1. Rationale
Largevolumeparacentesis(>5litres)associatedwithplasma volume expansion has been shown to be more effective than diuretictherapy in eliminating asciticfluid and wasassociated withalowerincidenceofcomplications[155].Inpatientstreated with paracentesis without plasma expansion, paracentesis-in-ducedcirculatory dysfunction(PICD) maybepresent in 70% of cases.ThediagnosisofPICDisbasedconventionallyonanincrease in plasmarenin activityof>50%of thepre-treatment valueto above 4ng/mL/h on the sixth day after paracentesis. PICD is associated with an increased rate of recurrent ascites, the development of hepatorenal syndrome, hyponatremia and re-ducedsurvival[156].Arecentmeta-analysisassessing17 random-ised studies (1225 patients) provided evidence that albumin infusion(8g/Lofascitesfluidremoved)afterparacentesisisthe mosteffectivetherapyinthepreventionofPICD,ascomparedto alternative treatments, and reduced the odds of PICD by 61% (OR=0.39, 95% CI 0.27–0.55) [157]. The ability of albumin to reducethiscomplicationwasalsoshowninasubgroupanalysis versuseachoftheothervolumeexpanders tested(e.g.dextran, gelatin,hydroxyethylstarch,hypertonicsaline).Withalbumin,the odds of hyponatremia and mortality were reduced by 42% (OR=0.58, 95%CI0.39–0.87)and by36%(OR=0.64, 95%CI95% 0.41–0.98),respectively [157]. Thefindings oftwo otherrecent randomisedstudies[158,159]didnotaltertheconclusionsofthe meta-analysis[160].Anotherrecentmeta-analysiscontestedthe effectofalbuminonmortality[161];however,thisfindingwas
dependentontheinclusionoftwostudieswithunsuitablecontrols (no treatment in one and mannitol in theother). Ifthese two studies were excluded, the benefit of albumin on mortality remainedsignificantin thewhole cohortofpatients (OR=0.58, 95% CI 0.40–0.86) and in the subgroup of patients without hepatocellular carcinoma (OR=0.60, 95% CI 0.39–0.91). The administrationofalbumin(6to8gperLascitesfluidremoved) isrecommendedtopreventPICDafterlargevolumeparacentesis (>5litres)[162,163].Inpatientswhoseparacentesisproducesless than5Lascites,theriskofdevelopingPICDislow.However,EASL clinicalpracticeguidelinesstatethatthesepatientsshouldstillbe treated with albumin because of concerns about the use of alternativeplasmaexpanderssuchasDextran.
One controversial issue remains the dose of albumin that shouldbe administered, but only a few studiesaddressed this question. The results of an unblinded randomised pilot study showedin70 patientswithlow severitycirrhosis (meanMELD Scoreat16to17)thattreatmentwithahalfdoseofalbumin(4g perLversus8gperLascitesfluidremoved)waseffectiveandsafe inpreventingPICD[164].AsthePICDriskisgreaterwhenmore than8litresofascitesfluidareremoved,itseemspreferable to limit ascitesremoval ofless than8 litresduringa paracentesis procedure[165].Finally,althoughtherehavebeennostudies,itis recommendedthatthealbumininfusionshouldbedoneslowlyin order to prevent any potential cardiac overload promoted by preexistingcardiomyopathy.
R7.2–InpatientswithcirrhosishospitalisedintheICU,we suggestthatconcentratedalbumininfusionsshouldbeusedin theeventofspontaneousbacterialperitonitis(SBP)
GRADE2+,STRONGAGREEMENT
12.2. Rationale
Spontaneous bacterial peritonitis(SBP) is an acute bacterial infection of ascitic fluid common in patients with cirrhosis (prevalence: 10% to 30%). Diagnosis requires paracentesis, or samplingoftheperitonealfluidfromtheperitonealcavity.When fluid contains large numbersof white blood cells (neutrophils) (>250cells/mm3),theinfectionisconfirmed.
Sort et al. showed that treatment with combination of intravenous albumin (1.5g/kg on day 1 and 1g/kg on day 3) and antimicrobial therapy could reduce the incidence of renal impairment(10% versus33%; P=0.002) and death (22% versus 41%;P=0.03),ascompared toantibioticalone [166]. Sincethis publication,numerousstudiesaboutfluidresuscitationinseptic context affecting cirrhotic patients have been published. Six randomisedcontrolledtrials (RCTs)including577patients with cirrhosisandSBP(4studies)orothertypesofbacterialinfection (2studies), compared patients assigned toreceiveantibioticor antibiotics plus albumin (0.14 to 1.5g/kg on days 1 and 3) (5 studies) or antibiotics plus hydroxyethyl starch (one study)
[166–171]. None of these studies were double-blinded and
included patients hospitalised in ICU. We performed a meta-analysis including additional data [172] showing that when comparedtostandardantibiotictherapyalone,treatment involv-ing albumin plus antibiotics improved survival at 3 months (OR=0.66; 95% CI 0.45–0.96, P=0.03). Albumin has beneficial effects on renal function involving lower incidence of renal impairment(6RCTs:OR=0.46,95%CI0.30–0.71,P<0.001).Since noneof thesestudieshavebeenperformed inICUpatients, the levelof therecommendation islowered becauseit isbased on indirectevidence.
Inaddition,onlytheoldeststudies,whichincludedonlySBP, showedthatalbumininfusionssignificantlyimprovedsurvivalat 3 months (OR=0.36, 95% CI 0.21–0.61; P=0.0001) and renal impairment(OR=0.21,95%CI0.11–0.42;P<0.0001).Thesedata restricted the recommendation to patients with SBP (studies carried out between 1999 and 2009). The benefits of albumin infusions inpatientswithSBPcouldbeexplainedby haemody-namic effects after paracentesis, although theymay merely be duetotheseproducts’expandingproperties.Albuminisbelieved tobeeffectiveinpatientswithSBPbecauseofitsabilitytoimprove intravascularvolume(alteredbyparacentesis)and tobind pro-inflammatory molecules.However, theidealdosehasyettobe determinedand analbumininfusionshouldbeassociatedwith haemodynamicmonitoring(volumestatus,cardiacoutput).
IncirrhoticpatientswithaninfectionotherthanSBP,twoRCTs failed to show that albumin infusion improved the survival, althoughitdelayedtheonsetofrenalfailure[169,170].Theuseof HES should be proscribed in patients with cirrhosis. The administration of HES may favour liver failure, particularly in cirrhosis setting. The rational for these adverse effects is the lysosomalstorageofHESinKupffercellsandhepatocytes. 13. Question8:Inpatientswithcirrhosishospitalisedinthe ICU,whichmanagementofacuteuppergastrointestinal bleedingshouldbeinitiatedtoreducemorbidityand mortality?
R8.1.1–Inpatientswithcirrhosisandacuteupper gastrointesti-nal bleeding, we recommend the introduction as soon as possible of intravenousvasoactive therapy (with octreotide, somatostatinorterlipressin)andprophylacticantibiotictherapy.
GRADE1+,STRONGAGREEMENT
R8.1.2–Inpatientswithcirrhosisandacuteupper gastrointes-tinalbleeding,wesuggesttheuseofprotonpumpinhibitorsas soonaspossible.
GRADE2+,STRONGAGREEMENT
R8.1.3–Inpatientswithcirrhosisandacuteupper gastrointes-tinalbleeding,werecommendtheperformanceofanupper endoscopyassoonaspossible.
GRADE1+,STRONGAGREEMENT
13.1. Rationale
Splanchnic vasoconstrictor agents such as somatostatin, octreotide and terlipressin exert their vasoactive effects and decreasesplanchnic blood flowand portalpressure. Inpatients with acute variceal bleeding who are undergoing endoscopic sclerotherapy,severalstudiesshowedthattheearlyintravenous administration of a vasoactive agent was more effective than placebointheoverallcontrolofhaemorrhage[173–175].Inone meta-analysis published in 2012 [176], the use of vasoactive agentswasassociatedwithasignificantlylowerriskofmortalityat 7days (RR 0.74; 95% CI 0.57–0.95; P=0.02),fewer transfusion requirementsandashorterdurationofhospitalisation.This meta-analysisincluded19studieswithseveralvasoactivemedications (octreotide, n=9; somatostatin, n=3; terlipressin, n=4; vaso-pressin,n=2;andvapreotide,n=1).Inamulticentre,randomised trialthatincluded780patients,theproportionoftreatmentfailure withinthefirst5days,wasnotdifferentwiththreewell-known
vasoactive drugs (octreotide, somatostatin or terlipressin) as adjuvanttherapytostandardendoscopictreatments[177].
Bacterialinfectionsareacommoncauseofmortalityinpatients with cirrhosis and upper gastrointestinal bleeding. A meta-analysisincluding12 controlled trials concludedthatantibiotic prophylaxiswasassociatedwithasignificantreductioninoverall mortality(RR0.79,95%CI0.63–0.98),inmortalityfrombacterial infections(RR0.43,95%CI0.19–0.97)[178]andamarked reduc-tioninoverallre-bleedingepisodesamongpatients under anti-bioticprophylaxis(RR0.53,95%CI0.38–0.74).Trialstoevaluate length of hospital stay have shown that patients receiving antibioticprophylaxistendedtobeadmittedforshorterperiods
[178].Inthemeta-analysisofthesetrials,twelvestudiedantibiotic
prophylaxisusingquinolones(fivetrials),quinolonesplus beta-lactams (two trials),cephalosporins (three trials),carbapenems (onetrial) and non-absorbableantibiotics (onetrial) versus no interventionor a placebo. Allthe antibiotics tested achieved a beneficialeffectonthebacterialinfection.
Onestudyfoundthatahigh-doseinfusionofaproton-pump inhibitorbeforeendoscopyacceleratedtheresolutionofsignsof bleedinginulcersandreducedtheneedforendoscopictherapy
[179].Thisreaffirmsthatoptimalacidsuppressionfacilitatesclot
formationoverarteriesinbleedingpepticulcers.Fewerthan5%of the patients included in this study had cirrhosis. However, becausethebleedingisnotrelatedtoportalhypertensionin30% ofpatientswithcirrhosis,theexpectedbenefitofproton-pump inhibitorspriortoendoscopyinthesepatientsledtheexpertsto overstatethelevelofevidenceforproton-pumpinhibitorsinthis indication.
Uppergastrointestinalendoscopyenablestheexaminationand also the treatment of most upper gastrointestinal bleeding. A combinationofbandingligationandvasoactivetherapyfor2days wassuperiortotheinfusionofvasoactivetherapyfor5daysalone in reducing very early re-bleeding [180]. Per-endoscopic treat-mentsusuallyconsistofbandingligationofoesophagealvarices andgluetherapyforgastricvarices.Patientswithupper gastro-intestinal bleeding and features suggestive of cirrhosis should undergoendoscopy within12hours ofpresentation [180–182]. Thisissupported bydatafroma non-randomisedstudy,which reportedlessrecurrenceofhaemorrhageandbettersurvivalwhen endoscopywas performed withinthefirst 12hours[183].This periodalsoallowsforarapiddiscussionontheadvisabilityofTIPS intheeventofvaricealhaemorrhageuncontrolledbyendoscopy anddrugtherapy(refractorybleeding).
R8.3–Inpatientswithcirrhosisandacuteupper gastrointesti-nalbleeding,wesuggesttoconsiderTransjugularintrahepatic portosystemicshunt(TIPS)withacoveredstentwithin24to 72hoursofthebleedingforepisodesinChild–PughclassC patients(<14)orinChild–PughclassBpatientswithinitially activebleedingatendoscopy(earlyTIPS).
GRADE2+,STRONGAGREEMENT
13.2. Rationale
ArandomisedcontrolledtrialevaluatedtheearlyuseofTIPSin cirrhoticpatientsinChild–PughclassCorclassB[182].Atotalof 63patientswithacutevaricealbleedingwereincluded.Thisstudy showedthattheearlyuseofTIPSwasassociatedwithasignificant reduction in mortalityat 2years. However, this studysuffered from several limitations that deserve consideration: the main endpointwastherateoffailuretocontrolbleedingwithin5days; seventeenpatientsweretreatedwithsclerotherapy,whichisnot the recommended endoscopic treatment, and the recruitment delay was long as 359 patients were screened and 63 were
includedbetween2004and2007.Thetwootherstudiesdidnot haveacontrolleddesign(lowlevelofevidence)[187,188].Another oldrandomisedcontrolledstudyspecificallyanalysedearlyTIPSin this indication,but withanuncoatedstentand theselection of high-risk patients, using the hepatic venous pressure gradient
[189].Finally,ameta-analysisassessedtheeffectsofearlyTIPSon
patient prognosis, but it was of poor quality because of large heterogeneitybetweenthetypesofstentsused(coatedornot)or theselectioncriteriaforpatients[190].
R8.4–Inpatientswithcirrhosisandacuteupper gastrointesti-nal bleeding, the experts suggest considering emergency Transjugularintrahepaticportosystemicshunt(TIPS)witha coveredstentintheeventofvaricealbleedingrefractoryto endoscopictreatment(salvageTIPS).
EXPERTOPINION,STRONGAGREEMENT
13.3. Rationale
Inastudyofpatientswithcirrhosisandvaricealhaemorrhage uncontrolledbyendoscopyanddrugtherapy,TIPS(salvageTIPS) wasabletohalt thehaemorrhagein 80%ofcases[191].Inthis settingofactivebleeding andafailure ofstandardmedicaland endoscopic haemostasis, alternative measures include balloon tamponade (Blakemore or Linton tube) or a self-expandable oesophagealstent.Boththesetreatmentscouldberecommended asabridgetodefinitive therapy(TIPStherapy)inpatientswith cirrhosisandmassiveorrefractoryoesophagealvaricealbleeding. There has only been one randomised controlled trial, which included 28 patients to compare self-expandable oesophageal stents and balloon tamponade in the treatment of acute and refractorybleedingfromoesophagealvarices[192].Therewereno significant differences in 15-day survival rates between the oesophagealstentandballoontamponadegroups.Theavailability ofoesophagealstentsislimited.Consequently,theexpertswere notabletomakearecommendationabouttheiruse.
14. Question9:Inpatientswithcirrhosishospitalisedinthe ICU,whichmeasurestomanagehaemostasisshouldbe initiatedtoreducemorbidityandmortality?
R9–Theexpertssuggestthatroutineprophylacticfreshfrozen plasma,platelets or fibrinogenconcentrates shouldnot be systematically administrated before invasive procedures in cirrhoticpatientsinordertoreducebleeding.
EXPERTOPINION,STRONGAGREEMENT
14.1. Rationale
Althoughconventionalcoagulationtestsarealteredinpatients withcirrhosis,thebalancebetweenpro-andanticoagulantfactors isrespectedandthegenerationofthrombinisnormal[193,194]. Post-proceduralbleedingisnotpredictedbythesetests(platelet count or international normalised ratio) [195]. Nor is there consensusregardingtheplateletorplasmatransfusionthresholds or any evidence of their clinical efficacy [196]. The French MedicinesAgency(Agencenationaledese´curite´ dume´dicament etdesproduitsdesante´)andtheEuropeanSocietyofHepatology (EASL) recommend an infusion of platelets when the count is<50G/Lbutthishasnotbeensubstantiatedbyanypublication
[197,198].
Standarddosesoffreshfrozenplasmararelycorrect coagulo-pathyinpatientswithcirrhosis.Toachievethat,largeamountsof
plasmaneedtobetransfused,associatedwithadverseeffectssuch asincreasedvascularvolumeandportalpressure,ornonspecific bloodproductsadverseeffects[199].
Althoughthe optimalfibrinogen level is uncertain, the pre-ventiveadministrationoffibrinogenisalsoamatterof consider-abledebate.A benefit/risk assessmentshould beperformed on a case-by-case basis, depending on the patient’s haemostatic parameters.
Useofathrombopoietinreceptoragonist(eltrombopag)could raisetheplateletcountandavoidtheuseofplatelettransfusionsin cirrhoticpatientspriortoaninvasiveprocedure.However,other reportshavesuggestedanincreasedriskofthrombosis.Therisk/ benefitratioisthereforeunfavourable[200].
Fourrandomised controlledtrials [201–204],and one meta-analysis [205] assessed the effects of rFVIIa prophylaxis on preventing mortalityand bleeding resulting fromhepatobiliary surgery(liver resectionor livertransplantation).There wereno significantdifferencesbetweenrFVIIaandplacebowithrespectto mortality(OR 0.96;95%CI0.35–2.62),redbloodcellunits(MD 0.32;95%CI 0.08to0.72)oradverseevents(OR1.55;95%CI0.97– 2.49)[205].
No recommendation: theexperts were not able to produce recommendationsconcerningthromboprophylaxisincriticallyill patientswithcirrhosis.
14.2. Rationale
Despite their low levels of coagulation factors, thrombin formation and clot formation are normal in cirrhotic patients
[193,206].Thesepatientshaveafrequentimbalancebetween
pro-coagulantandanticoagulantactivity,whichfrequentlyresultsina hypercoagulable state [207,208]. Several studies have demon-strated an increase in thrombotic events in cirrhotic patients versus a control population [209,210]. Very few studies have examined the value of thromboprophylaxis in patients with cirrhosis;thelevelofevidenceislowandonlybasedonone meta-analysisof non-randomisedretrospective series[211].The het-erogeneityofthestudiesinthis meta-analysisdidnotallowto formallyconcludeastotheeffectivenessofthromboprophylaxisto preventvenousthromboembolism(pooledOR0.8795%CI0.34– 2.18)[211].Inthesamemeta-analysis,thromboprophylaxiswas not associated with a higher rateof bleeding in patients with cirrhosis. Other retrospective non-randomised cohort studies confirmednoextrariskofbleeding.Onlyoneretrospectivestudy including 256 cirrhotic patients showed an increase in overall haemorrhagesbut not an increasedrisk of majorhaemorrhage
[212,213].
Prophylactictreatmentwithunfractionatedheparinhasbeen showntobeassociatedwithanincreasedriskofbleedingevents
[214]. We recommend assessing the individual risk of venous
thromboembolism,particularlyinpatientswithcirrhosiswhoare undergoingsurgery.Forpatientsathighriskofthromboembolism andwithahighbleedingrisk,useofanintermittentpneumatic compressiondeviceasprophylaxisforvenousthromboembolism maybeanalternative[215,216],butithasnotbeenevaluatedin patientswithcirrhosis.
15. Question10:Inpatientswithcirrhosishospitalisedinthe ICU,whenshouldexpertadvicebeobtainedinordertoreduce morbidityandmortality?
R10–Theexpertssuggestthatexpertadviceshouldbesought regardingpatientswithcirrhosiswhoarehospitalisedintheICU: 1. At admission, if the patient is already on the liver transplantationwaitinglist.
2.SoonafterICUadmissiontodiscusstheintensityofcaretobe providedasafunctionofthenumberoforganfailuresandtheir course.
3.Inordertodiscussthebenefitsofaliversupporttechnique. 4.AtdischargefromtheICU inordertoorganisehepatology management,withaviewtopotentialpossibleliver transplanta-tion
EXPERTOPINION,STRONGAGREEMENT 15.1. Rationale
1.Theonsetofcomplicationsinpatientswithcirrhosiswhoare on thelivertransplantwaitinglistshouldbereportedtotheLT centre.Indeed,acutedecompensationdoesnotcontraindicateliver transplantationandmightevenfavouritbyprioritisingthepatient onthelist.
2.TheneedforICUadmission,whateverthereason,isassociated with high mortalityratesin patientswith cirrhosis(ICUand 6-monthmortalityratesofrespectively42.7%and75%)[114].The prognosisforthesepatientsdependsontheseverityofbothhepatic failure (evaluated using the MELD score at ICU discharge) and extrahepaticorganfailures(respiratory,neurological, haemodyna-mic,renal,coagulation,hepatic)[104,114,217].Dependingonthe numberoforganfailures,28-daymortalityraterangesfrom4.7%in patientswithoutorganfailureto76.7%inpatientswithACLFgrade 3 [68].Evolutionofthe SOFAscoreand/or thenumberof organ failures during the ICU stay is a crucial prognostic factor. A retrospective study that included 138 patients with cirrhosis supportedtheconclusionthatthemostimportantriskfactorfor in-hospitalmortalitywasthemodifiedSOFAscore(without haema-tologicalfailure)assessedafter3days[69].Inanotherstudy,the SOFA score proved to be the most accurate, with a score >10 predictiveof mortality in93% of cases[90]. However,the deltaSOFAscore(differencebetweenbaselineandat48hor3-day) seemstodisplayvariousprognosticvalueswhencomparedtothe scorevalueat48hor3-daywithrespecttoshort-termmortality
[69,90,217].Persistenceof3or4non-haematologicalorganfailures
at 3-day [69] or between 3-day and 7-day [29–220] predicted inhospital mortalitywithaspecificity of93%or100%.Thesame results were observed regarding the use of organ replacement methods(vasopressors,mechanicalventilation,renalreplacement therapeuticorartificialliversupportsystems)duringthestayof patients in intensive care [77,115]. In the CANONIC study, assessment of theACLF gradevariationbetweenJ3and J7after diagnosiswassignificantlybetterabletopredict28-dayand90-day mortalityratesthantheACLFgradeatdiagnosis(AUROC0.86versus 0.65and0.81versus0.62P<0.0001,respectively)[220].
Prognosisassessmentforcirrhoticpatientsadmitted inICUis moreaccurateifperformedwithanalysisofthenumberoffailing organs,ortheSOFAscoreafewdaysafterICUadmissionthanonICU admission.Thenumberoforganfailuresmustbetakenintoaccount whenidentifyingapatientinwhomaggressivetreatmentmayoffer recoveryor determinecandidacyfortransplantation,in orderto optimisecare[69,116,221].Inrecentyears,livertransplantationin patients with ACLF has been considered to be feasible. Retro-spectives studies that included patients with ACLF grade >2showedthatlivertransplantationimprovedthepoorprognosis of the most severely ill patients with cirrhosis [220–229]. The probabilityofsurvivalwas78%atoneyearinpatientsreceivingan earlytransplantcomparedtolessthan10%inthosenottransplanted
[222–229]. However, in a context of organ shortage, individual
beneficeshouldbeconsideredtogetherwiththegeneralinterest, sinceanincreasedpostoperativemortalityhasalsobeenreported aftertransplantationforthemostseverepatients.Numerousstudies havetriedtoidentifyfactorspredictiveofearlypost-LTmortalityin ordertoavoidfutileLT.Thus,severalspecificscoringsystemsbased
